Understanding spatial variation in the driving pattern of carbon dioxide emissions from taxi sector in great Eastern China: evidence from an analysis of geographically weighted regression

Abstract

Transport is a significant sector that contributes to the increasing carbon dioxide emissions in China. Most related studies have utilized global models to investigate the factors influencing transport sector carbon dioxide emissions from the provincial level to the national level. However, this approach cannot depict the spatial nonstationary characteristics of the influencing factors. Taking the taxi subsector in Eastern China as a case study, we employed exploratory spatial data analysis to examine the spatial heterogeneity of carbon dioxide emissions from taxis (CDET) at the city level and a local model (geographically weighted regression, GWR) to observe the spatial variation in the factors influencing emissions. The results showed that the spatial distribution of CDET was uneven across the study region; the provincial-level cities and metropolises had the highest level of carbon dioxide emissions. The variables gross domestic product (GDP), total population (TP) and urban built-up area (UBA) were among the most significant determinants of emissions. Moreover, using the GWR technique instead of conventional global models, we successfully detected variation in the relationships between emissions and factors at the city level. The GWR results revealed that the positive correlation between CDET and GDP was distributed throughout the study area in 2005 and 2014, but the value of the regression coefficient did not remain the same across the study area. Both positive and negative effects of TP and UBA on CDET were identified and mapped in both study years. Based on the results, we strongly recommend that differentiated mitigation measures be adopted to reduce CDET emissions for different cities according to the spatial variation in emissions and the factors that impact them.